Tissue Hydration Monitor

Abstract

A tissue hydration monitor and method includes a sensor module having a plurality of LEDs positioned to emit light of a plurality of different wavelengths toward the user's skin and a detector detecting light transmitted and reflected by the user's skin to generate a signal corresponding to the intensity of the light detected at each different wavelength. A processor / controller module generates a baseline hydration level based on the received signals, calculates a relative hydration level, and generates an output indicating relative hydration personalized to the user. The housing is secured against the user's skin by an adhesive patch or strap.

Description

[0001] Related Applications

[0002] This application claims the benefit of priority to U.S. Provisional Application No. 62 / 663,926, filed on April 27, 2018, which is incorporated herein by reference in its entirety. Field of the Invention

[0003] The present invention relates to tissue hydration monitors and, more particularly, to tissue hydration monitoring systems and methods for personalized hydration monitoring. Background of the Invention

[0005] Dehydration is a condition in which the amount of water in an organism is reduced to a level below which the individual can function normally. Dehydration occurs when an individual exercises for an extended period of time without taking in enough water to offset internal water losses through breathing, perspiration, and waste removal, or when temperatures rise to a point where the individual cannot produce enough sweat to maintain their normal body temperature. People who frequently exercise in low humidity and / or high temperature conditions and / or for extended periods of time are susceptible to experiencing dehydration or symptoms of dehydration. The elderly and children are also particularly susceptible to experiencing dehydration or symptoms of dehydration.

[0006] In less severe cases of dehydration, an individual’s ability to perform tasks may begin to deteriorate. For example, in the case of an endurance athlete or professional athlete, a person who becomes dehydrated by losing 2% of their body weight may begin to experience a decrease in performance. Low levels of hydration can lead to hypovolemia, which impairs circulation, nutrient exchange, hormone balance, and waste removal. In dehydration, sodium levels in the blood decrease, leading to hyponatremia, the initial symptoms of which include fatigue, headache, weakness, and nausea. Additional manifestations include cramping, disorientation and confusion, swelling of the limbs, and, in extreme cases, swelling of the brain. Cramping is common in athletes and is a good key indicator that the body is depleted of electrolytes. Losing more than 5% of body weight can reduce a person’s ability to complete a task by 30%.

[0007] There are currently no wearable sensors on the market that provide accurate and personalized hydration requirements. It is important for athletes to have a self-calibrated measurement of hydration needs, as each athlete has specific needs based on their diet, exercise, and genetics. In order to maintain optimal performance, it is necessary to monitor an individual's hydration level regularly or even continuously so that fluctuations can be detected at an early stage before performance levels are affected, especially before a person approaches a critical dehydration condition. The present invention is directed to a method and apparatus that allows a person to easily monitor and provide notification of hydration levels to allow early correction before dehydration occurs.

[0008] Brief Overview

[0009] In an exemplary embodiment, the wearable optics device uses a light emitting diode to emit light and one or more photodiodes to detect light from the skin. In a preferred embodiment, the device will use at least 4 wavelengths of light (e.g., 740nm, 850nm, 940nm, 1450nm) and a single detector. Based on LED availability and technical adjustments, other wavelengths can be used, such as variations of + / -10nm. Ideally, the LED and photodiode will be at least 1 cm apart from each other, with a light-blocking partition in the middle to prevent any light from traveling directly from the LED to the detector without penetrating the skin. The depth of light propagation in the tissue depends on the wavelength, source-detector spacing, and tissue composition, and can be up to several centimeters.

[0010] The wearable device includes a wireless communication module (e.g., or similar wireless communication circuitry) and corresponding software to communicate with a smartphone, tablet, or computer that has stored an application ("app") for inputting settings and for receiving and storing data.

[0011] In one aspect of the invention, the system controller collects information from the user over a predetermined time period to establish a personalized baseline from which subsequent performance is measured. The thresholds for generating alarms or other deviation indications may be preset by the system controller, or the user may use an associated app to tighten or expand the threshold ranges. In one embodiment, the system controller or an associated app on the user's phone, tablet, or computer may execute a learning algorithm or other appropriate algorithm to continuously or periodically update the user's personalized baseline. For example, as the user's condition improves, or as the intensity of exercise increases, he or she may experience changes in hydration response that should be considered when determining acceptable ranges.

[0012] The device can be equipped with a strap to allow the device to be worn on the wrist, upper or lower arm, ankle, calf, or other location on the limb. The strap should be flexible and elastic enough to press the sensor area firmly against the skin to prevent external light from entering the sensor and ensure that the light-blocking baffle effectively prevents light from the LED from being directly received by the detector without passing through the skin. In a preferred embodiment, a non-irritating, skin-friendly pressure-sensitive adhesive ("PSA") is used to temporarily fix the sensor to the skin. Such adhesives can be purchased from a variety of sources and are used to fix the skin-bonded device for a period of several hours to a week.

[0013] The present invention relies on near infrared spectroscopy of tissue. This approach is not new and has been extensively studied. Prior patent applications have been filed for measuring tissue hydration using near infrared spectroscopy. However, the prior art uses different wavelengths, different algorithms, or both, and does not provide a personalized calibration system for determining the optimal hydration range.

[0014] In one aspect of the invention, a tissue hydration monitor includes: a housing configured to be disposed against a user's skin; a sensor module disposed within the housing, the sensor module including: a plurality of LEDs configured to emit light toward the user's skin at a plurality of different wavelengths; and a detector configured to detect light transmitted through and reflected from the user's skin from each of the plurality of LEDs over a time period and generate a signal corresponding to the intensity of the light detected at each different wavelength; a processor / controller module configured to receive a signal from the sensor module, execute an algorithm for generating a baseline hydration level based on the received signal, calculate a relative hydration level at a time point within the time period, and generate an output indicating the relative hydration at the time point; and a power source configured to power the sensor module and the processor / controller module. The time period may include a period of user activity ranging from rest to exercise, wherein a baseline hydration range is generated at a time point during rest, and relative hydration levels are calculated at a plurality of time points during exercise. The plurality of different wavelengths are wavelengths absorbed by oxyhemoglobin, deoxyhemoglobin, lipids and water, and are preferably 740 nm, 850 nm, 940 nm and 1450 nm. An indicator responsive to the processor / controller module is provided for generating an alarm indicating a predetermined deviation from the baseline hydration level. The predetermined deviation is preferably a change of less than or equal to +2%. The indicator may be an LED arranged to emit light from a surface of the housing. A light shield is preferably provided around the detector to prevent stray light from impinging on the detector.

[0015] In some embodiments, an adhesive material is applied to the housing to removably seal the sensor to the user's skin. The adhesive material can be in the form of a patch sized to seal the perimeter of the housing against the user's skin to create a substantially light-tight seal. The patch can be formed of at least a breathable material and a waterproof material. In other embodiments, a strap can be attached to the housing for holding the housing against the user's skin.

[0016] The tissue hydration monitor may also include a communication module in communication with the processor / controller module for transmitting data to a remote mobile device for display output. The mobile device may have an application installed therein having instructions for further processing the output.

[0017] In another aspect of the invention, a method for monitoring hydration in a subject includes placing a sensor device against the subject's skin, the sensor device including a plurality of LEDs configured to emit light toward the user's skin at a plurality of different wavelengths; detecting light transmitted through and reflected from the user's skin from each of the plurality of LEDs over a time period and generating an intensity signal corresponding to each of the different wavelengths of light; generating a baseline hydration level based on the intensity signal; calculating relative hydration levels relative to the baseline hydration level at a plurality of time points over the time period; and generating an output indicating the relative hydration at the time point. The plurality of different wavelengths are wavelengths absorbed by oxyhemoglobin, deoxyhemoglobin, lipids, and water, and may be 740 nm, 850 nm, 940 nm, and 1450 nm. The method may include transmitting data to a remote mobile device for display output and generating an alarm indicating a predetermined deviation from the baseline hydration level, wherein the predetermined deviation is a change of less than or equal to +2%. The time period includes periods of subject activity ranging from rest to exercise, wherein a baseline hydration range is generated at time points during rest, and relative hydration levels are calculated at multiple time points during exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1A is a prior art graph showing the absorption coefficients of water, whole blood, fat, and melanin, with each tissue component having several different concentrations; Figure 1B Prior art graphs of the absorption coefficients of oxyhemoglobin, hemoglobin, water, and lipids are provided.

[0020] Figure 2 Depicted are how the oxygen saturation S and the relative hydration level ΔW may change over time and how they will manifest in the received light intensity.

[0021] Figure 3A is a graphical representation showing hydration versus dehydration measured over a five day period using water absorption according to the method of the present invention; Figure 3B is a graph comparing dehydration measurements obtained using water absorption and urine specific gravity over the same period.

[0022] Figure 4 is a block diagram of the basic components of an embodiment of the hydration sensor of the present invention.

[0023] Figure 5 is a block diagram showing a data processing flow of an embodiment of a sensor system according to the present invention, in which user interface components and background processing components are indicated.

[0024] Fig. 6A -6e shows a different embodiment of a wearable hydration monitor, wherein Fig. 6A is a schematic diagram of the sensor (front and rear views); Figure 6B and Figure 6C The underside and the upper side of a second embodiment of the present invention are shown; Fig.6D and Fig. 6E An alternative LED layout is shown.

[0025] Figure 7 is a graph of exemplary hydration data over time for a personalized hydration index.

[0026] Figure 8 is an exploded view showing an embodiment of the device of the present invention attached to the skin of a user.

[0027] Detailed description of embodiments of the present invention

[0028] refer to Figure 1A (from Jacques below), provides absorption coefficients for water, whole blood, fat, and melanin for different wavelengths from ultraviolet to infrared, with several different concentrations of each tissue component. Figure 1B A graph of the absorption coefficients of the major endogenous agents in biological tissue is provided. Light in the first two wavelength ranges, approximately 740 nm to approximately 760 nm (red) and approximately 850 nm to approximately 940 nm (NIR), will provide measurements of deoxygenated hemoglobin (Hb) and oxyhemoglobin (HbO2), respectively. Using these two measurements, we can determine tissue oxygen saturation from known absorption data. See, e.g., SL Jacques, “Optical properties of biological tissues: a review,” Phys. Med. Biol. 58 (2013) R37-R61, Pellicer and Bravo Mdel, “Near-infrared spectroscopy: a methodology-focused review,” Semin Fetal Neonatal Med., Feb 2011; 16(1): 42–49. doi: 10.1016 / j.siny.2010.05.003, both of which are incorporated herein by reference for background information on optical properties of tissues.

[0029] The Fourier transform of the 850nm signal will provide the user's heart rate. At 940nm (and also at about 1200nm and 1420nm), lipids are the main absorbing chromophores, so this signal will provide a measure of the lipid content of the tissue. Starting from about 1000nm, with significant peaks at about 1480nm and 1950nm, water becomes the main absorber, providing a measure of the water content of the tissue. In an exemplary embodiment, light at 1450nm (NIR) is selected to detect water absorption, primarily for economic reasons. Using light with each of the four discrete wavelength ranges will provide an indication of scattering and melanin contributions together.

[0030] The intensity of the received light over the time and range scale of interest can be expressed by the absorption coefficient μ a and the scattering coefficient μ s It is approximated by a linear combination of .

[0031] I=A1μ a +A2μ s (1)

[0032] Where A1 and A2 are unknown constants and depend on the sensor geometry, electronics, and other factors that do not vary significantly over the time frame of interest. Absorption coefficient μ a There is a complex dependence on wavelength and tissue composition. Within the wavelength of interest, the scattering coefficient can be approximated as a linear function of wavelength.

[0033] μ s =s1+s2λ. (2)

[0034] Equation 3 provides the total absorption coefficient for typical tissues:

[0035] μ a =BSμ a,oxy +B(1-S)μ a,deoxy +Wμ a,水 +Fμ a,fat +Mμ a,黑素体 +2.3C bili ε bili +2.3 βC ε βC (3)

[0036] Where S is the HGb oxygen saturation of the mixed arteriovenous vascular system, and B is the mean blood volume fraction (f v.血液 ), W is the water content (f v.水 ), Bili is the bilirubin concentration (C(M)), βC is the β-carotene concentration (C(M)), and F is the fat content (f v.脂肪 ), M is the volume fraction of melanosomes (f v.黑素体), or alternatively, the concentration of melanin monomers (C(M)).

[0037] Each of these terms may depend on time and the wavelength of the light. Over the time range of interest, the last four terms do not vary significantly with time, but rather with the wavelength of the light, so they can be lumped together into a single parameter, D. Again, we are primarily interested in estimating the relative change in hydration (W) from the baseline (starting) value, so W(t i ) is divided into W1+ΔW(t). Writing this out with explicit dependence on the time wavelength λ and time t,

[0038] μ a (λ,t)=

[0039] BS(t)μ a,oxy (λ)+B(1-S(t))μ a,deoxy (λ)+(W1+ΔW(t))μ a,水 (λ)+

[0040] D(λ). (4)

[0041] The device of the present invention will measure four wavelengths (with λ k ) and thousands of time points (denoted by t i Represents the intensity of light at a location.

[0042] I(λ k ,t i )=A1(BS(t i )μ a,oxy (λ k )+B(1-S(t i ))μ a,deoxy (λ k )+(W1+ΔW(t i ))μ a,水 (λ k )+D(λ k ))+(s1+s2λ k ). (5)

[0043] S is the level of HGb oxygen saturation and will change with each arterial pulse on a time scale of one second or less. The relative hydration ΔW will change on a time scale of minutes or hours, and our goal is to track these changes. A variety of signal processing techniques can be used to measure the relative hydration in the presence of other unknown quantities A1, A2, S(t i ), W1, D, s1 and s2 to estimate ΔW(t i ). An example is to use a device to record all I(λ) over a period of several minutes or hours. k ,t i) data for one or more least squares fits (regressions). If appropriate, I(λ k ,t i ) is used to separate the spectral components of I.

[0044] The relative hydration value can be used to generate a personal hydration index ("PHI") for the user. In an exemplary implementation, for an initial set of measurements (i.e., "training data"), the first 3 minutes of the training phase are used to establish an average intensity value for the sensor's four LEDs. Data is then collected for the duration of the training phase. At the end of the training phase, the average intensity value for the entire phase will serve as the baseline for the next phase. The notification indicating "out of optimal range" can initially be pre-set at a threshold of 2% from the baseline, based on the general consensus in the scientific literature that such deviations correspond to the onset of cognitive deficits. In a preferred embodiment, the notification will be generated before a change of + / - 2% occurs to allow the user sufficient time to rehydrate before an out-of-range condition is reached. During a given training or activity phase (i.e., a monitored time period), the device of the present invention will periodically measure and calculate relative hydration to determine the user's relative hydration level at multiple time points, where the monitored time period can include activity periods ranging from rest to exercise. The type of exercise varies from person to person and can range from walking to intense athletic activity. At any given time point during the monitored time period, if the measured relative hydration approaches or exceeds a predetermined threshold, the device of the present invention can generate a notification to prompt the user to rehydrate. The notification can occur at multiple time points during exercise, especially during long and intense activities.

[0045] Ideally, the notification would be generated before the user's performance could be affected by their hydration status. In prototype testing, a sensitivity of 0.1% was achieved, which is 10 times more sensitive than thirst (1-2%). With more data and more advanced analytics, this range could be personalized to allow the user to set his or her own ideal threshold for allowing active hydration, thereby maintaining the user's optimal hydration level throughout the activity. In other embodiments, signal processing techniques such as machine learning and data mining can be employed to derive I(λ k ,t i ) and ΔW(t i ) to generate the user’s PHI and appropriate notification thresholds.

[0046] Figure 2It is shown how the oxygen saturation S (upper panel) and the relative hydration level ΔW (middle panel) may vary over time and how they will appear in the received light intensity (lower panel), for example, at wavelengths of 700 nm (for HbO2), 850 nm (for Hb), 1200 nm (for lipids) and 1500 nm (for water).

[0047] Figure 3A and Figure 3B is a graph of water absorption measurements taken over a five-day period. Figure 3A Shown are the normalized water absorption levels of the test subjects one hour after waking up (dehydration) and drinking 473 ml (16 oz) of water (hydration) each day. During the one-hour period, the subjects performed light activities and ate breakfast. In each case, the measurement duration was 5 minutes. The values ​​measured during the collection period were averaged and then normalized, with hydration used for normalization. Figure 3B Dehydrated water absorbance measurements using the 1450 nm signal were compared to urine specific gravity (USP) measurements of urine collected on waking (dehydrated) and showed good correlation. The line at USP 1.01 is the delineation where any measurement above USP 1.01 is considered dehydrated.

[0048] First reference Figure 4 4, wherein there are three main modules in the sensor assembly 402: a sensor module 406, a processor / controller module 422, and a wireless communication module 424. In the sensor module 406, LEDs 409-412 (i.e., LED 1, LED 2, LED 3, and LED 4) can be controlled to pulse on and off sequentially, while a detector 415 continuously detects to measure the intensity of reflected light from each LED. Power for the components within the sensor assembly 402 is provided via a battery 418, preferably rechargeable via a USB or similar charging port 416. In some embodiments, an inductive charging circuit may be included, such as that widely used in smart watches and some smart phones. An on / off switch 417 may be provided to enable manual control of power consumption. The collected data will be processed by the processor / controller module 422, the results of which may be stored in a memory 423 and then communicated via (or similar wireless communication module 424) to a mobile application 432 downloaded to a smartphone 430, tablet, computer, or other personal portable device. The user will be able to use the mobile application 432 ("app") to interact with the device and determine his or her current hydration status as well as review past measurements. An indicator light (LED) 420 is connected to the processor / controller module 422 to provide quick visual feedback to alert the user when a preset condition (e.g., dehydration or overhydration) is detected. Alternatively, or in addition to the indicator light 420, the sensor device may include an audio or vibration device connected to the processor / controller module 422 for generating an audible or tactile alarm to indicate that the user's hydration level is no longer within the desired optimal hydration range.

[0049] Figure 5 A flow chart of the functionality within the various components of the monitoring system of the present invention is provided. In the illustration, elements with solid outlines indicate user interfaces, while dashed outlines indicate background processing. The sensor assembly 402 is activated after being placed in close contact with the user's skin. The LED is activated and a light signal is detected and processed within the processor / controller module 422. The detected signal 502 is transmitted via wireless transmission 504 to a mobile device 430, which has previously been loaded with a mobile app 432. In step 506, the app 432 performs processing within the mobile device 430, including establishing a baseline range for the user's personal hydration index ("PHI") based on relative hydration and normalized intensity using the above-mentioned algorithm, and then comparing the corresponding value of the current detected signal to the PHI to determine whether the user is dehydrated, overhydrated, or within his or her optimal hydration range (step 510). Brief Reference Figure 7 , a sample graph of exemplary hydration data over time for a personalized hydration index (PHI) determined through an initial calibration period is shown. In the graph, the y-axis is the PHI, with a baseline optimal hydration range indicated in the shaded area and activity-induced changes in the user's hydration level relative to the optimal range indicated along the x-axis.

[0050] return Figure 5In one embodiment, in step 512, determining that the user is within the optimal range may result in sending a message to the sensor assembly 402 to activate a green display at the indicator light 420, while being out of range may generate a red or yellow display. In another embodiment, step 512 may activate a different color for each of underhydration, overhydration, and optimal hydration, for example, blue light for overhydration, red for underhydration, and green for optimal hydration. The yellow display may provide an additional indication that the user is approaching the limit of their optimal range. As described above, step 512 may alternatively or additionally result in the generation of an audible or tactile alarm. In step 516, the determined hydration level may be sent to a remote data collection / storage medium for, for example, community data mining. To provide an example of a possible application of this step, data may be collected by a coach or team trainer who may monitor the athletic performance of many team members on the sidelines, providing additional data for decisions regarding player substitutions or developing insights into the impact of hydration on team performance. In step 518, the mobile app may be updated to include current data, allowing historical records to be stored.

[0051] The device combines novel hardware and software. Figures 6A-6E As shown, in an exemplary embodiment, the housing 404 of the wearable sensor 402 is a flat disc, typically having a diameter of 25-60 mm (about 1-2.5 inches) and a thickness of about 8-12 mm (about 0.3-0.5 inches). As technological advances allow for further miniaturization of sensor components, the dimensions may be modified. Although the shape of the housing 404 is shown as being roughly circular in cross section, other shapes may obviously be used. For example, an ellipse and a polygon with sufficiently large corner angles may be used to avoid sharp corners that may cause skin irritation or injury if the device is impacted during athletic activity. In addition, although the profile of the housing 404 is illustrated as being roughly flat (having a uniform thickness over the entire diameter), the periphery of the housing may taper toward the edge to produce a flat convex profile. The edges of the housing should be inclined or rounded to reduce the edges that may injure anyone who contacts the device during athletic activity. Silicone-based (non-irritating, non-allergenic) adhesives can be used to close and seal housings to protect device electronics from moisture intrusion from sweat and other liquid exposure. Figure 4 As discussed, the hardware enclosed within the housing includes a battery 418, LEDs 409-412, 420, a photodiode (light detector) 415, a micro USB port 416, a switch 417, a voltage regulator, a microcontroller, an analog front end (all in a processor / controller module 422), and Or similar wireless communication module 424.

[0052] Battery 418 and 424 will allow the device to operate and communicate wirelessly with the mobile application 432. The micro USB port 416 will allow for battery charging and wired data transfer. The voltage regulator, microcontroller, analog front end will power and control the sensor. The battery life must be sufficient to drive the LEDs, data collection, indicators, and Bluetooth data transmission for at least 24 hours.

[0053] Fig. 6A An embodiment of a wearable sensor is shown that utilizes an adhesive fabric patch 440 to removably secure a sensor 402 to a user's skin to ensure close and continuous contact for extended periods of time and / or during activity. A silicone-based adhesive system will keep the device in place and provide water resistance. Skin-friendly adhesives are commercially available for attaching sensors and similar devices for periods ranging from a few hours to approximately a week. The adhesive should be able to stay on the skin and measure comfortably for at least 24 hours. In some embodiments, the adhesive can be a liquid or gel that can be applied directly to the skin using an applicator or brush. The patch should be soft enough, such as foam, gel, fabric, or a combination thereof, to avoid skin irritation. The adhesive fabric patch 440 has a larger diameter than the sensor housing so that the entire perimeter of the patch contacts and adheres to the user's skin to provide a light-tight and waterproof seal around the sensor. A small opening 426 in the patch 440 allows for viewing of the indicator light 420. The patch can be, for example material, with a non-irritating adhesive, pre-cut to the correct size to achieve the desired coverage, or a kinesiology tape, such as a KT tape (KT ), can be cut to size to hold the sensor in place. The extended tabs 442 in the pre-cut patch provide an easy grip feature to facilitate application and removal of the patch.

[0054] Figure 8An example of applying the sensor to the user's skin using a pre-cut multi-layer adhesive patch is shown, which includes a dual adhesive middle layer 444 cut to match the outer shape of a patch 441 that contacts the outer surface of the sensor 402. In some embodiments, the outer patch 441 can be permanently attached to the outer surface of the sensor housing 404, with only the middle layer 444 being replaced, while in other embodiments, the outer patch 441 can be removable and replaceable, provided with the middle layer 444 as a package with a removable backing sheet that can be peeled off to expose new adhesive for application to the skin. The central opening 446 in the middle layer 444 fits tightly around the edge of the sensor 402 to provide a light-tight seal around the device, thereby preventing external light from being received by the detector 415. The middle layer 444 (facing the skin) can be formed of a water-absorbing material to absorb sweat generated by exercise, thereby minimizing the impact of signal artifacts due to moisture. The outer patch 441 of the adhesive system is preferably permeable to airflow to prevent overheating and allow the skin to remain dry. The design of the adhesive patch is important to the performance of the sensor in terms of moisture and light control. In general, whether a single-layer or multi-layer patch, it should be able to keep the sensor in place for 24 hours and limit movement between the skin and the device, while allowing the skin to remain dry enough to minimize signal artifacts caused by sweat.

[0055] In an exemplary implementation, light sources 409-412 are LEDs selected to emit light at each of 740 nm, 850 nm, 940 nm, and 1450 nm for detecting the absorbance of oxygenated hemoglobin (HbO2), deoxygenated hemoglobin (Hb), lipids, and water, respectively. The specific combination of wavelengths selected is unique because they together provide comprehensive personalization of measurements including fat content, skin color, tissue oxygenation, heart rate, and water content. It should be noted that the detection target exhibits absorption peaks at other wavelengths, and the selection of appropriate combinations of LEDs and wavelengths will be within the technical level of the art. In general, any wavelength combination within the range of 600-2000 nm sufficient to provide blood flow, oxygen saturation, lipids, and water content can be used. In one embodiment, the wavelength of the LED can be changed to target the spectrum of other or additional chromophores of interest. Different algorithms can also be applied to normalize and process the data to obtain similar / improved results. Various classification algorithms can also be used to achieve similar / improved results.

[0056] refer to Fig. 6A, LEDs 409-412 are located in the plane of the contact surface 405 of the sensor to at least partially surround the detector 415 and are equidistant from the detector 415. A light protection ring 425 surrounds the detector 415 to prevent light from the LEDs from directly entering the detector from the sides, so that only light reflected upward by the skin shines on the detector 415. Because the protection ring 425 is preferably pressed firmly against the user's skin to form a good seal to prevent light intrusion, it should be formed of a soft, pliable material, such as a silicone O-ring or a similar compressible, non-irritating material.

[0057] The housing 404 will preferably be formed of an opaque (light-proof) plastic or polymer material, such as black or dark, to minimize stray light reflected or transmitted. The housing may be formed of two halves that fit together and seal to minimize moisture intrusion into the interior of the device. In some embodiments, the housing 404 may be a rigid material, a rigid material with an elastic coating (for comfort and / or increased safety), or an elastic material with sufficient rigidity to provide protection for the electronic device components encapsulated therein. It may be desirable to fill the interior of the housing with a potting material or other protective agent to ensure that the electronic device components are well protected from moisture and impact. Each of the LEDs 409-412, indicator lights 420, and detectors 415 will preferably be located behind a transparent window that is sealed to the interior of the housing to create a waterproof seal. The window may be partially recessed in the surface, i.e., not flush with the contact surface 405, to serve as a light shield to minimize lateral light leakage into the detector.

[0058] With adhesive patch fastener Fig. 6A and Figure 8 Embodiments provide an advantage in that the sensor can be affixed anywhere on the user's body that has a sufficiently flat surface area to achieve a good seal. Figure 6B and Figure 6C An alternative method of securing a hydration sensor to a user's skin is shown, which is typically applied to a limb or body part that can be worn with a belt or strap. In this embodiment, the sensor housing 404 is attached to a strap 450 that can be wrapped around the user's arm or leg. Figure 6B As shown, strap 450 may be fastened using a hook-and-pile fastener (e.g. ) to connect, and Figure 6C , the strap is shown as including a buckle 454 attached to the strap. In either variation, the strap 450 will preferably be elastic enough to firmly but comfortably hold the sensor against the user's skin with minimal slippage.

[0059] Fig.6D and Fig. 6EAlternative arrangements of LEDs and detectors are shown. In each case, the LEDs are positioned to at least partially surround the detector 415, which is generally located near the center of the skin contact surface 405 of the housing 404. The positioning of the LEDs relative to the detectors is designed to aid in the identification of motion artifacts by signal processing methods. Other arrangements following the same general configuration will be apparent to those skilled in the art.

[0060] The data processing algorithm described above is used to generate a personal hydration index (PHI) that takes into account light scattering, melanin, lipid content, and blood flow as well as water content. Another parameter that can be measured and incorporated into the algorithm is heart rate. The algorithm can use the heart rate frequency with water content to measure arterial water content. When water is consumed, it is first absorbed from the stomach into the blood stream and then distributed to skeletal muscle. Lipid content can be used to normalize the PHI for athletes with different body mass indexes (BMI). Testing may be required to determine the maximum BMI at which the device may not be effective in providing accurate measurements due to a thicker lipid layer. The algorithm that uses heart rate to isolate and ensure that water measurements are from arterial blood is novel. The algorithm that uses individualized scattering coefficients as well as lipid content to normalize the data is novel. The algorithm that combines all measurements into a personalized hydration index is novel.

[0061] When in use, the user should preferably place the device on an area of ​​skin that directly overlies a large artery and has a sufficiently flat and smooth area, such as no folds, bends, wrinkles, or prominent scars, to allow the sensor to lie flat on the skin. Possible locations include the brachial artery on the inner side of the biceps, the posterior tibial artery on the back of the calf, and the femoral artery on the inner side of the thigh. Fig. 6A , Figure 6B and Figure 8 As described, the device can be applied using a double-sided adhesive patch that can be left on the body for at least 24 hours. After cleaning the area, the user can reuse the device by replacing the adhesive patch to continue monitoring their PHI.

[0062] Motion artifacts can be a major problem with light detection on the skin. Methods to minimize these motion artifacts include ensuring that the adhesive is applied to keep the sensor on the skin without detaching. Using a water-absorbing material on the adhesive patch will help absorb sweat that accumulates during exercise. For embodiments that employ an adhesive patch, the patch should be formed of a breathable material to allow airflow and minimize moisture accumulation between the sensor and the skin. The LED pulse frequency can be set to be semi-random to avoid repetitive frequencies that may cause motion artifacts (e.g., heart rate). By using a moving average of the data points, the data can be smoothed and large artifacts due to exercise movement can be excluded.

[0063] The calibration period used to establish a personal baseline should preferably be at least 12 hours, during which the upper and lower limits of the PHI are calculated. Preferably, the calibration procedure should be performed on a day when the user's activity is relatively low to allow for a more accurate determination of the optimal hydration range. The LEDs may also be turned on and off at different frequencies to avoid synchronizing the heart rate with any motion artifacts (heartbeats).

[0064] Another approach to establishing a baseline PHI may include employing a hybrid procedure: a long collection period (12 hours), or using a predetermined "standard PHI" and then continuously updating the baseline with each successive workout to gradually personalize it for the user. Using the PHI, the relative change in hydration during the workout cycle can be determined.

[0065] In some embodiments, for example, where a hydration monitor is used on a team of athletes, data can be collected for monitoring by a coach or trainer, with identification information intact to allow the coach or trainer to identify individual athletes who may need to be called out of an activity for rehydration. In other embodiments, the data can be stripped of identification information and stored in the cloud for the purpose of data mining. Each user can be tagged with one or more tags that classify the user's sports groupings. For example, tags can include, but are not limited to, selected exercises, resting heart rate, height, weight, gender, race, exercise duration, exercise intensity, sports classification (e.g., amateur or professional), or other characteristics that can be used to distinguish factors that may affect hydration and / or sensitivity to hydration fluctuations.

[0066] Using these tags, various data mining algorithms can be used to classify athletes as overhydrated, dehydrated, or optimally hydrated based on their PHI. Possible data mining algorithms include learning machines (neural networks, support vector machines, Bayesian networks, genetic algorithms, etc.), statistics, clustering, regression, etc. The computational processing can be done locally on the mobile application, or on a cloud computing system, where the measurement results can be compared with people belonging to the same or similar classification based on data accumulated from other users and stored in a central database. Data analysis can include feature selection and machine learning algorithms to identify the attributes that best determine hydration. Once classified, one or more indicator LEDs can be activated to, for example, display red for measurements that are outside the optimal range, green for those that are within the optimal range, and yellow or orange for insufficient data / errors or close to being outside the optimal range. Users can access more detailed information through the mobile application.

[0067] The more data collected, the more accurate the PHI will be, and the more robust the classification algorithm will become to indicate dehydration, overhydration, or optimal hydration. The measurements can be improved by applying more than one sensor on the body. By taking measurements in different areas of the body, an overall whole-body average hydration can be generated, rather than point measurements. This technique may help discard anomalous data due to artifacts in one location.

[0068] Using more than one sensor would require more than one adhesive patch (or strap) and a modified data processing algorithm to combine the measurements and tease out artifacts. A combined whole-body personal hydration index (PHI) may be calculated from multiple measurement sites.

[0069] In some cases, a user may not want to take continuous real-time measurements during activity. In these cases, there is no need to use an adhesive or strap. Instead, the user can simply place the sensor on dry skin for a continuous measurement period, such as before and after a workout, to provide two measurement points. Assuming calibration to the optimal hydration range has been achieved, this simple measurement provides data on how much water loss has occurred during a given exercise session. A similar approach can be used if the user wishes to monitor his or her hydration during rest or other breaks in activity. To provide a practical argument, it is not uncommon for an athlete to suffer a severe muscle cramp during a soccer or tennis match. Taking a quick measurement during a break in play can allow an athlete to quickly determine if they are close to dehydration and take preventative measures before an actual cramp occurs.

[0070] The software includes a self-calibration algorithm that uses the high and low hydration ranges from at least one calibration day (24 hours) before the device is ready to accurately determine the personalized optimal range, taking into account tissue scattering, lipid content, and water content. Once the optimal range is determined, the device will be able to determine the percentage of PHI change, indicating overhydration, dehydration, or optimal hydration during exercise. The more measurements the user records, the more accurate the optimal range will be, and therefore the more accurate the PHI prediction will be.

[0071] The user may decide to use only this or her own measurements, where processing may be handled in the mobile app, or he or she may choose to use cloud computing techniques known in the art to access data from other users on the cloud. Ideally, once enough data is loaded onto the cloud, the athlete will be able to compare and contrast his / her exercise level and hydration needs with other similar athletes.

[0072] Novel aspects of the inventive apparatus and method include specific pre-processing of the data to feed the classification algorithm, and the option to run the classification locally on the mobile application or using cloud computing.

[0073] Although the present disclosure has been shown and described with reference to one or more implementations, equivalent substitutions and modifications will occur to others skilled in the art upon reading and understanding this specification and the drawings. The present disclosure includes all such modifications and substitutions and is limited only by the scope of the appended claims. In particular, with respect to the various functions performed by the above-mentioned components (e.g., elements, resources, etc.), unless otherwise indicated, the terms used to describe these components are intended to correspond to any component (e.g., functional equivalent) that performs the specific functions of the described components, even if they are not structurally equivalent to the disclosed structures that perform the functions of the exemplary implementations of the present disclosure shown herein.

Claims

1. A tissue hydration monitor, comprising: a sensor module configured to be disposed within a housing having a contact surface, the housing configured to be secured against the skin of a user; The sensor module comprises: a plurality of LEDs configured to sequentially emit light pulses from the contact surface toward the user's skin through one or more light-transmissive windows disposed internally, each LED emitting at a discrete wavelength corresponding to one of a plurality of chromophores of interest; and a detector configured to sequentially detect pulses of light emitted by each LED, wherein the light is transmitted into the user's skin, absorbed by the user's skin portion, and reflected over a period of time, and to generate a signal corresponding to an intensity of the detected light at each wavelength, wherein the detected light is received via the one or more light-transmissive windows; a processor / controller module configured to receive signals from the sensor module, execute an algorithm for generating a baseline hydration level based on the received signals, calculate a relative hydration level at a point in time within the time period, and generate an output indicative of the relative hydration at the point in time; and a power supply configured to provide power to the sensor module and the processor / controller module; Wherein, the plurality of chromophores of interest include oxygenated hemoglobin, deoxygenated hemoglobin, lipids and water.

2. The tissue hydration monitor of claim 1, wherein: The wavelengths corresponding to the plurality of chromophores of interest include a first discrete wavelength in the range of 740 nm to 760 nm, a second discrete wavelength in the range of 850 nm to 940 nm, a third discrete wavelength selected from 940 nm, 1200 nm and 1420 nm, and a fourth discrete wavelength selected from 1000 nm, 1450 nm, 1480 nm and 1950 nm.

3. The tissue hydration monitor of claim 1 further comprising a communication module in electrical communication with said processor / controller module for transmitting data to a remote mobile device for displaying said output.

4. The tissue hydration monitor of claim 3, wherein: The mobile device has an application installed therein, the application including instructions for further processing the output.

5. The tissue hydration monitor of claim 1 further comprising an indicator responsive to said processor / controller module for generating an alarm indicating a predetermined deviation from said baseline hydration level.

6. The tissue hydration monitor of claim 5, wherein: The indicator includes an LED arranged to emit light from a surface of the housing.

7. The tissue hydration monitor of claim 5, wherein: The predetermined deviation is a change of less than or equal to +2%.

8. The tissue hydration monitor of claim 1, further comprising a light shield disposed around the detector to prevent stray light from shining on the detector.

9. The tissue hydration monitor of claim 1, wherein: The time period includes periods of user activity ranging from rest to exercise, wherein a baseline hydration range is generated at time points during rest, and relative hydration levels are calculated at multiple time points during exercise.

10. A method for monitoring hydration in a subject, the method comprising: placing a sensor device against the skin of the subject, the sensor device comprising a plurality of LEDs configured to sequentially emit pulses of light, each LED emitting at a discrete wavelength corresponding to one of a plurality of chromophores of interest; sequentially detecting light transmitted and reflected by the user's skin from each of the plurality of LEDs over a time period and generating an intensity signal corresponding to the light at each of a plurality of discrete wavelengths; generating a baseline hydration level based on the intensity signal; calculating relative hydration levels relative to the baseline hydration level at a plurality of time points during the time period; as well as generating an output indicative of relative hydration at the time point; Wherein, the plurality of chromophores of interest include oxygenated hemoglobin, deoxygenated hemoglobin, lipids and water.

11. The method according to claim 10, wherein: The plurality of discrete wavelengths include a first discrete wavelength in the range of 740 nm to 760 nm, a second discrete wavelength in the range of 850 nm to 940 nm, a third discrete wavelength selected from 940 nm, 1200 nm and 1420 nm, and a fourth discrete wavelength selected from 1000 nm, 1450 nm, 1480 nm and 1950 nm.

12. The method of claim 10, further comprising transmitting data to a remote mobile device for display of the output.

13. The method of claim 10, further comprising generating an alarm indicating a predetermined deviation from the baseline hydration level.

14. The method according to claim 13, wherein: The predetermined deviation is a change of less than or equal to +2%.

15. The method according to claim 10, wherein: The time period includes periods of subject activity ranging from rest to exercise, wherein a baseline hydration range is generated at time points during rest, and relative hydration levels are calculated at multiple time points during exercise.

16. The method of claim 10, further comprising generating a personal hydration index (PHI) by: calculating an average hydration intensity value of the plurality of LEDs during an initial calibration period after hydration of the subject; Collect intensity values ​​throughout the calibration period; normalizing the collected intensity values ​​using the average hydrated intensity value; The average normalized collected intensity values ​​were calculated to establish a baseline personal hydration index, PHI.

17. The method according to claim 16, wherein: The personal hydration index PHI is updated with each consecutive activity period.

Images